Current issues of ACP Journal Club are published in Annals of Internal Medicine


Review: Initial clinical assessment is accurate for determining causes of dyspnea

ACP J Club. 1993 Nov-Dec;119:90. doi:10.7326/ACPJC-1993-119-3-090

Source Citation

Mulrow CD, Lucey CR, Farnett LE. Discriminating causes of dyspnea through clinical examination. J Gen Intern Med. 1993 Jul;8:383-92.



To determine whether history and physical examination are useful in distinguishing the causes of dyspnea.

Data sources

MEDLINE searches for 1966 to 1991 and review of bibliographies from major textbooks and relevant articles using the terms dyspnea, differential diagnosis, decision making, diagnostic, heart disease, sensitivity and specificity, predictive value, and decision were done; experts were consulted; and relevant citations were tracked.

Study selection

Studies were selected if the patients had dyspnea as a presenting symptom, if the study was not limited to a single diagnosis, if bedside maneuvers were used to determine the cause of dyspnea, and if > 5 patients were studied.

Data extraction

3 authors independently rated all studies on quality (5 levels). Criteria and prevalence of dyspnea, setting, etiologies, inclusion and exclusion criteria, and diagnostic tests used were collected. When possible, diagnostic accuracies and operating characteristics were extracted or calculated.

Main results

251 studies were identified, 68 were reviewed in detail, and 11 were included, with 3 assessed as high quality (1 or 2 on a 5-level scale). Overall accuracies of the initial clinical assessment (including basic laboratory tests) ranged from 66% to 92% with 1 high-quality study having 74% accuracy. One high-quality study evaluated items from the history and physical examination. Dyspnea with exertion had a likelihood ratio of 1.3 (95% CI 1.0 to 1.6) and orthopnea had a likelihood ratio of 2.0 (Cl 1.0 to 3.9) for the diagnosis of left ventricular (LV) dysfunction. The blood pressure response to the Valsalva maneuver (forced expiration against a closed glottis or external airway obstruction) was evaluated in 2 high-quality studies. The sensitivity, specificity, and likelihood ratio for identifying patients with dyspnea who had heart failure or LV dysfunction were 88%, 90%, and 8.8 (CI 2.3 to 33.2) for a study of 37 patients with obstructive lung disease, cardiac disease, or both; and 78%, 65%, and 2.1 (Cl 1.0 to 4.2) for consecutive patients in an emergency department. The abdominojugular reflux maneuver (AJR) was evaluated in 4 studies. In the highest quality study, the AJR identified congestive heart failure with a sensitivity of 24%, a specificity of 96%, and a positive likelihood ratio of 6.4 in unselected patients with dyspnea.


The initial clinical assessment is accurate for determining the cause of dyspnea. Valsalva and abdominojugular reflux maneuvers are useful for determining dyspnea of cardiac origins.

Source of funding: Not stated

For article reprint: Dr. C.D. Mulrow, Ambulatory Care (11C), Audie L. Murphy Memorial Veterans Hospital, 7400 Merton Minter Boulevard, San Antonio, TX 78284, USA. FAX 210-567-4685.


In the autobiographic novel, The Intern (1), a young physician recounts a near disaster when he incorrectly treats a patient with emphysema for pulmonary edema. This is a common story. We all struggle with the differential diagnosis of dyspnea. Unfortunately, the technical support that can help with diagnosis is not often available when fundamental decisions must be made. For better or worse, we must rely on the history and physical examination to make these decisions.

The study by Mulrow and colleagues provides valuable information. The most important finding was the value of the history in establishing previous probabilities for the physical examination. A history of previous disease increased the predictive value of certain physical findings (e.g., crackles correlated with interstitial lung disease in patients with a previous history of obstructive lung disease). Similarly, a history of dyspnea after exertion, of orthopnea, and of paroxysmal nocturnal dyspnea was helpful in the diagnosis of congestive heart failure.

For physical findings, jugular venous distension and gallops were not helpful, nor were pulmonary wheezes and crackles. 2 special tests proved valuable: the Valsalva maneuver and the AJR test. The authors point out the problems inherent in interpreting such data when the patient cannot cooperate, which is often the situation when dyspnea is the problem.

Future studies will show us how to make better use of the data collected in the history and physical examination and how to develop more practical models for diagnosis (2) to help with initial decisions.

Thomas A. Parrino, MD
Brown UniversityProvidence, Rhode Island, USA


1. Dr X. The Intern. New York: Harper and Row, 1965.

2. Hull RD, Raskob GE, Carter CJ, et al. Pulmonary embolism in outpatients with pleuritic chest pain. Arch Intern Med. 1988;148:838-44.